***REPLICATION ARCHIVE FOR: Campaign Mailers and Intent to Turnout: Do Similar Field and Survey Experiments Yield the Same Conclusions? (Doherty and Adler)***

This replication archive includes two datasets: field_experiments_replication_data.dta (data as provided in replication archive for Doherty and Adler, 2014) and mturk_replication_data.dta (data from the survey experiment). 

The file Replication_Code.do executes all analysis reported in the paper. The first portion handles recoding of the field experiment data. It also stacks observations from the two experiments and identifies which observations are tied to which experiment. As part of this process, data from the first experiment are saved to the "scratch_datasets" folder. The stacked field experiment data is also stored to this folder so it can be appended to the survey experiment data. 

It then opens the data from the survey experiment and drops all variables not used in the analysis reported in the paper. Variables used are then recoded as needed (see Online Appendix for details regarding coding of the data). 

Tables and figures are saved to the (empty) "tables" folder.

Before executing the replication code, the user should create two empty folders in directory that holds the data files: 1) tables; 2) scratch_data